Under the July 26, 2026 True Up rules, a Cisco Services EA only beats standalone SmartNet when your hardware support estate stays above roughly 85 percent of committed value for all three years and clears the $200k ACV floor without discount give-back
Cisco's shift from True Forward to True Up on hardware support removes the two features that made the EA a hedge: forward-only billing and Partial Commit. What remains is a non-reducible Full Commit floor of $200k ACV with retroactive annual invoicing, which flips the comparison toward a-la-carte for any estate that shrinks, refreshes, or migrates during the term.
Prepared by Redress Compliance · September 3, 2026 · Cisco advisory practice. Services EA and SmartNet renewal engagements, 2024 to 2026.
Executive summary
The EA discount premium on hardware support has to clear a $200k ACV floor and a Full Commit obligation before it is worth anything, and most mid-market estates cannot clear both.
Cisco's own FAQ confirms that customers below $200k ACV are simply not eligible to include HW CX in a Services EA, and that the threshold bites only when at least one HW CX suite is in the proposal, so the gate is entirely self-inflicted by the deal structure your account team proposes.
True Up converts every growth event into a retroactive invoice at the annual milestone, erasing the timing benefit that justified the EA premium for a decade.
Cisco marketed True Forward for years by explicitly contrasting it with agreements "that require a true-up every year that charges you for past use," and on July 26, 2026 it adopts exactly that model for hardware support while leaving software support on True Forward.
Cisco lists "decreasing discounts" as an approved route to reach the $200k floor, which converts a compliance threshold into a price increase mechanism.
The Services EA Customer FAQ names five remedies for customers who fall below the floor at renewal, and one of them is simply paying more for the same coverage, so a buyer who accepts the framing without pushback can absorb a double-digit uplift on flat volume.
Standalone SmartNet keeps three rights the EA now removes: line-item decommission, multi-partner bidding, and no cross-suite entanglement.
EA 3.0 requires the Services portfolio to be bought through the same partner as the product portfolio, kills Partial Commit on hardware, and makes commitments non-reducible except under narrow decommission conditions.
So an a-la-carte estate of the same size carries materially lower structural risk even at a worse headline discount.
How the two cost models actually calculate: Services EA versus a-la-carte SmartNet
The two models are not variations on a theme, they are different financial instruments.
A Services EA hardware line is a committed annual contract value: you and Cisco agree a number, that number is billed for the full term, and from July 26, 2026 it is Full Commit only, because Partial Commit has been withdrawn from hardware services.
Growth above the commit is reconciled at the annual milestone and billed retroactively under True Up, so consumption that started in month two of the anniversary year generates eleven months of back charges on the invoice.
Reduction runs the other way with no symmetry: Cisco's own portfolio documentation states payment obligations are based on your initial coverage and may only be reduced when you decommission assets under defined conditions such as a technology refresh.
Standalone SmartNet is the opposite instrument: a per-serial, per-year coverage line, priced against list for each device, cancellable or non-renewable at the individual contract line, and re-quoted every renewal cycle. There is no floor, no milestone, no commit.
The tradeoff you are buying with the EA is discount depth and price protection against per-device renewal drift, and the tradeoff you are selling is the right to shrink.
Read the entry mechanics alongside our breakdown of the $200k hardware and $100k software thresholds before you model anything, because eligibility, not price, is what decides most of these deals.
| Mechanic | Services EA hardware support | Standalone SmartNet |
|---|---|---|
| Entry threshold | $200k minimum ACV, enforced when any HW CX suite is in the proposal | None |
| Commit type | Full Commit only from July 26, 2026 (Partial Commit withdrawn) | Per-serial, no commitment |
| Growth billing | True Up, retroactive to the point of consumption, invoiced at annual milestone | Priced at the moment of purchase, forward only |
| Shrink rights | Non-reducible except on qualifying decommission or technology refresh | Cancel or non-renew at contract line level |
| Growth allowance | None on hardware (the 115% allowance covers Collaboration and Security software only) | Not applicable |
| Price protection | Original order price holds as ceiling for the term | Re-quoted every renewal, exposed to list and uplift changes |
| Partner choice | Services must be bought through the same partner as the product portfolio | Any authorized reseller, re-bid annually |
| Exit | End of term, or transition hardware coverage out of the EA at renewal | Any anniversary, per line |
The table cannot show the one genuinely valuable thing the EA still buys: Full Commit price protection, where your original order price becomes the ceiling for the entire term.
On a growing estate that is real money, because a-la-carte SmartNet is re-quoted per renewal and exposed to list uplifts, discount erosion, and the per-device repricing that follows every hardware refresh. The catch is that a ceiling only pays out when you press against it.
If your device count is flat or falling, you have paid a commitment premium for protection against a price rise on volume you never bought.
The break-even: where the EA discount stops covering the commit risk
Run the arithmetic honestly and the EA case narrows fast.
In the deals we negotiate, the incremental EA discount over a well-negotiated a-la-carte SmartNet position on the same install base typically lands in the low double digits, not the 30 or 40 points buyers assume from the software side of the house (that range is our field experience.
Not a published Cisco figure).
Call it a 15 point incremental advantage. That discount is earned only on coverage you actually consume, but the commit is paid on 100 percent of committed value.
So the EA wins only while utilized ACV divided by committed ACV exceeds roughly one minus the incremental discount, which at 15 points puts the break-even at about 85 percent utilization, sustained across all three years, not averaged.
Below that line you are buying a discount on coverage you are not using, which is arithmetically identical to paying full price for less.
| Sustained utilization of committed HW ACV | Effective cost vs a-la-carte at 15 pt EA discount | Verdict |
|---|---|---|
| 100% | 15% cheaper | EA wins clearly |
| 90% | ~5.5% cheaper | EA wins on a thin margin |
| 85% | Roughly break-even | Coin flip, terms decide it |
| 75% | ~13% more expensive | A-la-carte wins |
| 60% | ~42% more expensive | EA is a material loss |
A single 15 percent estate reduction, one campus consolidation, one branch refresh onto smaller platforms, one workload migration to cloud, drops you from 100 to 85 and erases the entire discount for the remaining term.
The asymmetry compounds it: overconsumption is billed retroactively at the milestone, so growth costs you back-dated months, while underconsumption is never refunded, credited, or carried forward.
You cannot use Value Shift to escape either, because Cisco's EA 3.0 program terms are explicit that a shift may not decrease the overall EA commitment.
Model your worst realistic three-year hardware trajectory, not your base case, and read our True Up versus True Forward financial comparison before you accept a commit number.
Cut your Cisco collaboration cost across the Webex stack
How to cut Cisco collaboration cost across Webex Suite, Calling, and Contact Center: the Flex Plan math and the contract levers that hold at renewal.
Get the white paper →Analysis: Cisco has moved the risk of estate volatility from its balance sheet to yours
Read the July 26, 2026 change as a risk transfer, not a pricing action. Under True Forward, Cisco absorbed the uncertainty of not knowing how fast a customer would grow: you consumed above entitlement, and Cisco billed the increment forward from the milestone, forgiving the period of underpayment.
That forgiveness was the product. Cisco was, in effect, extending free credit against its own forecasting problem, and it priced that credit into the suite discount. Under True Up, the same growth generates a retroactive invoice at the annual milestone.
The uncertainty being priced is now yours, and the credit runs the other way: you fund Cisco's coverage of assets from the day they attach, then settle at the milestone. Nothing about the underlying service changed. The party carrying the timing exposure did.
The split treatment is the tell. Partial Commit survives on software support and dies on hardware. If Cisco believed hardware support attach was healthy, it would defend the attach rate through product mechanics, not through a contractual floor.
Removing the ability to commit a subset of the hardware estate means Cisco no longer wants to compete for the marginal switch or router at renewal, because refresh cycles, cloud migration.
And the drift toward Meraki and subscription-based operating models are shrinking the hardware base faster than the software base.
When a vendor stops competing for the increment and starts locking the floor, it is telling you which line item is eroding. See our reading of the Partial Commit removal on hardware for the mechanical consequences.
Cisco has also paid a real reputational price here, and that is worth pricing into your read.
For years its own True Forward FAQ drew an explicit contrast with vendors that "require a true-up every year that charges you for past use," selling growth "without a retroactive, surprise bill." Cisco has now adopted the model it spent a decade positioning against, in one portfolio, on one date.
Vendors do not reverse published differentiation for marginal gain. They do it when the revenue line under the differentiation is deteriorating faster than the marketing value of defending it.
In 25 years of negotiating with this vendor, a reversal of that visibility has almost always preceded further tightening in the same portfolio within two renewal cycles.
The $200k ACV floor should be read as segmentation, not as a compliance rule. It does no protective work: Cisco explicitly preserves the right to buy hardware support a-la-carte below the line. What it does is sort the base.
Everyone under $200k is pushed into standalone renewals, where list-based uplift and per-contract renewal mechanics apply cleanly and no aggregate discount protects the whole estate.
Everyone sitting just above the line is pushed into a defensive purchase, buying suites they did not need, upgrading tiers they did not scope, adding Professional Services, or accepting a lower discount, simply to preserve eligibility. Both outcomes raise yield.
Neither requires Cisco to win a technical argument.
That is why the eligibility test is more dangerous than the invoice mechanics. A True Up bill for growth is at least proportionate to consumption.
Falling three percent below the floor at year three is not proportionate to anything: it either costs you the EA construct entirely or costs you a discount give-back applied across the full committed value. The floor is a cliff, and cliffs are where vendors extract.
The buyer's inference is simple. Combine a non-reducible Full Commit, retroactive milestone billing, no Growth Allowance on hardware services, and a hard eligibility floor, and the Services EA has become a three-year bet on your own hardware forecast.
Almost no infrastructure organization has a forecast of that quality.
Most cannot tell you within fifteen percent what their supported chassis count will be in month thirty, because refresh timing, budget deferrals, site consolidation, and platform migrations are all decided outside the network team.
The practical test is not "is the EA discount bigger than the a-la-carte discount." It is "how confident am I that my supported hardware value stays above 85 percent of commit in all three years.
And above $200k ACV at renewal." If you cannot defend both numbers in front of your CFO with a documented asset plan, you are not buying a discount, you are underwriting Cisco's revenue floor with your own budget.
Treat every EA discount point above your a-la-carte baseline as the premium Cisco is charging you to take that forecasting risk off its books. If the premium is four or five points, it is not adequate compensation for a non-reducible commitment with a cliff at renewal.
The five remedies Cisco offers below the floor, ranked by what they cost you
Cisco's FAQ lists five ways to reach $200k ACV at renewal, plus a sixth path it also confirms: move hardware support out of the EA. Rank them by cost per incremental dollar of coverage actually consumed, not by headline discount. Only two of the six deliver value you would have bought anyway.
The $200k and $100k threshold mechanics matter here because the floor is tested only when at least one hardware suite sits in the proposal, which makes exit a live option rather than a concession.
| Remedy | Real cost per $1 of useful coverage | Ranking and read |
|---|---|---|
| Transition hardware support out of the EA | At or near $1, plus loss of aggregate discount | Best default. Cisco explicitly permits it. You keep per-contract renewal levers and can time coverage to asset life. |
| Upgrade to a higher support tier | Roughly $1 if the SLA is genuinely required, otherwise pure waste | Least bad of the in-EA fixes. Only defensible where onsite response or parts SLA is a documented operational requirement. |
| Expand customer scope (more entities, sites) | Under $1 if the scope was already unsupported, above $1 if manufactured | Acceptable only when it formalizes coverage you are already buying elsewhere or self-insuring. |
| Add Professional Services | Highly variable, frequently unconsumed at term end | Risky. PS burns on a use-it-or-lose-it basis and inflates ACV without protecting hardware. |
| Buy more suites | Above $1 unless the suite was on the roadmap | Cisco's preferred outcome. Locks additional non-reducible commit to fix an eligibility problem. |
| Decrease discounts | Infinite: zero incremental coverage | Worst. A pure price increase dressed as an eligibility remedy. |
The item to fight is the last one. Cisco listing "decreasing discounts" as a legitimate path to compliance means an account team can present a five to eight point give-back as a technical requirement rather than a commercial ask. It is not.
It buys you nothing except the right to remain in a construct that has already stopped serving you.
If your estate is under the floor and shrinking, exit to a-la-carte, keep the renewal negotiations annual and contract-level, and revisit the EA only when a genuine multi-year hardware program restores the volume.
Structural terms that never show up in the price comparison
Every EA-versus-a-la-carte spreadsheet I have reviewed in the last two years compares discount percentages and stops there. The clauses that actually determine your three-year cost sit in the EA 3.0 Program Terms and the Offer Description, and they are uniformly asymmetric.
Start with the one buyers get wrong most often: Growth Allowance does not cover hardware services at all.
Cisco's own Offer Description (EDCS-25881943 v1.1) states the allowance applies only to associated Software Services within the Security and Collaboration portfolios, with "No Services on Hardware." So the 115 percent headroom (105 percent in the first six months.
115 percent thereafter) that your account team cites as flexibility is irrelevant to the very portfolio that carries the $200k floor.
Every additional chassis, switch, or firewall you add to hardware support is billable at the milestone, retroactively, with no buffer.
Value Shift is similarly narrower than it sounds: it can offset unconsumed value against growth in another eligible suite, but it may not decrease the value of your overall EA commitment, and it only executes if you gave your EA Authorized Partner 60 days advance written notice.
And only where the Full Commit suites were bought from that same partner, at the same time, for the same term, on the initial order.
Miss the notice window by a day and the mechanism does not exist that year.
Three further terms deserve line-item negotiation.
Exceptional Growth gives Cisco the right to call an off-cycle billing event when consumption exceeds a defined percentage of then-current entitlement.
So the semi-annual anniversary becomes a second exposure date rather than an administrative checkpoint; if you cannot get the percentage defined numerically in the order form, treat the milestone as effectively floating and read our milestone date planning analysis before you sign.
The Services portfolio must be bought through the same partner as the product portfolio, which removes competitive bidding on the single largest line in the deal.
And Expert Care and Asset Management are barred from the EA entirely, meaning your "consolidated" support estate is not consolidated: you will run parallel paper regardless.
| Structural term | What Cisco's paper actually says | Your negotiation ask |
|---|---|---|
| Growth Allowance | Excludes all services on hardware | Written 10 percent hardware support headroom before True Up applies |
| 115 percent allowance | Collaboration and Security software only | Confirm in writing it is not a hardware hedge |
| Value Shift | Cannot reduce total commit; 60 days notice; same partner, same term, same order | Reduce notice to 30 days; allow post-signature suite additions |
| Exceptional Growth | Cisco may call an off-cycle event at semi-annual anniversary | Define the trigger percentage numerically; cap to one event per year |
| Partner lock | Services must follow the product partner | Split product and services partners, or price-benchmark before signature |
| Excluded services | Expert Care, Asset Management sit outside the EA | Price these standalone and subtract the "consolidation" benefit |
Read the table as a set of one-way ratchets. Every mechanism that could move money in your favor (Growth Allowance, Value Shift, partner competition) is either scoped away from hardware, procedurally gated, or contractually prohibited from reducing your commitment.
Every mechanism that moves money toward Cisco (True Up retroactive billing, Exceptional Growth, Full Commit only) operates automatically.
Practically: if you cannot win the Value Shift notice period, the Exceptional Growth definition, and a partner split, the structural terms alone are sufficient reason to keep hardware support a-la-carte and preserve annual re-pricing on your own schedule.
What we see in the field: patterns across recent renewals
Proposals routinely land within 3 to 8 percent of the threshold, which is engineering, not coincidence.
Cisco's Offer Description states plainly: "No Services on Hardware."
The behaviors repeat with enough consistency across recent renewals that we now treat them as a checklist rather than observations.
First, proposals engineered to just clear the floor: low-value suites and a Professional Services line get bundled in until ACV crosses $200k, and the bundle is presented as scope rationalization. Second, and this is the one that costs the most, discount give-back framed as an eligibility fix.
Cisco's own FAQ lists "decreasing discounts" as a legitimate route to meeting the threshold, so a five-point erosion arrives labeled as compliance rather than a price increase. Read the threshold mechanics before you accept that framing.
Third, milestone dates set to fall a quarter or two after a known refresh, so the newly installed estate is captured retroactively at full True Up value while the decommissioned gear still carries commit.
Fourth, partner lock on the services portfolio suppressing any competitive quote, which in our experience is worth 4 to 9 points of margin to the incumbent.
Fifth, and most common in year two, customers discovering that assets pulled out of service still sit in the committed base, because payment obligations track initial coverage and reduce only under narrow decommission conditions.
The timing lever is the cleanest remaining play. Renewing or booking before July 26, 2026 preserves True Forward and Partial Commit for the full term.
If your renewal falls within twelve months of that date, pull it forward and lock the old mechanics; our renew-before-July-2026 decision framework sets out the qualifying conditions.
- Percentile standing for your exact deal size and industry, from real closed transactions
- Scenario simulation before the call: test alternative terms and see the financial impact of each
- A negotiation playbook, talking points, and a two page executive brief on day one
Your first five moves
- Model committed ACV against a three-year decommission forecast before you sign any EA hardware line, because Cisco's own Services Portfolio Features language says payment obligations track your initial coverage and reduce only under narrow refresh conditions, so every switch you retire in year two is a line you keep paying for.
- Demand the a-la-carte SmartNet quote in writing as the formal comparator, priced on the same install base and the same term, since Cisco's FAQ explicitly preserves the right to buy HW CX outside an EA and you cannot defend a commit you never benchmarked against the alternative.
- Refuse discount reduction as a threshold remedy and force scope or tier changes instead, because Cisco lists "decreasing discounts" alongside expanding scope, adding suites, upgrading tier, and adding Professional Services as valid paths to the $200k floor, and only one of those five hands back margin you already won.
- If you are renewing near the deadline, test whether closing before July 26, 2026 preserves True Forward and Partial Commit on hardware, then price that option side by side using our read on whether to renew your Cisco Services EA before July 26, 2026, since forward-only billing plus a partial hardware commit is worth real money on a volatile estate.
- If your hardware support spend sits below $200k ACV, default to a-la-carte and negotiate multi-year price caps on SmartNet directly, rather than buying suites or upgrading tiers purely to clear the $200k hardware and $100k software thresholds, and get renewal uplift ceilings in the paper before the contract vehicle debate is settled.
Frequently asked questions
What exactly changes on July 26, 2026 for Cisco Services EA hardware support?
From that date, all new and renewed Services EA bookings move hardware support (HW CX) from the True Forward model to True Up. True Forward billed growth going forward only; True Up generates a retroactive invoice at the annual milestone for growth that already occurred.
Software support (SW CX) stays on True Forward, so a single EA can now run two different billing models side by side.
Can I still buy Cisco hardware support outside an Enterprise Agreement?
Yes. Cisco's Services EA Customer FAQ explicitly confirms that customers who do not meet the $200k ACV requirement can still purchase HW CX outside of an EA.
Standalone SmartNet or a-la-carte HW CX remains a fully supported purchasing route and is the correct default for estates that cannot clear the floor without buying coverage they do not need.
Is the $200k threshold annual contract value or total contract value?
For hardware support it is $200k annual contract value (ACV). Software support and Professional Services carry a $100k total contract value (TCV) platform-level requirement instead, which is a materially lower bar on a three-year term.
Confirm which figure your account team is quoting, because conflating ACV and TCV is the most common way an underqualified deal gets presented as eligible.
Does the 15 percent growth allowance protect me on hardware support?
No. The EA 3.0 Offer Description states that Growth Allowance applies only to associated Software Services within the Security and Collaboration portfolios, and explicitly excludes services on hardware. The 115 percent entitlement headroom that Cisco markets is a software concept.
Every unit of hardware support growth is billable at the milestone.
If my estate shrinks mid-term, can I reduce the Services EA commitment?
Only in narrow circumstances. Cisco's terms state payment obligations are based on initial coverage and may only be reduced if you decommission assets under certain conditions such as a technology refresh.
Value Shift can move value between suites but cannot decrease overall EA commitment, and it requires 60 days advance notice to the EA partner before an annual event.
Why did Cisco keep Partial Commit for software but remove it for hardware?
Cisco's stated rationale is that Full Commit on hardware guarantees a minimum ACV, enterprise-wide coverage, multi-suite discounts, and Value Shift eligibility.
The commercial reading is different: hardware support attach is the more erodible revenue line under refresh cycles and cloud migration, so Cisco has locked the floor there while leaving software flexible. Treat the removal as a revenue defense, not a customer benefit.
Does renewing before July 26, 2026 protect me?
For the term of that renewal, yes. Bookings completed before the effective date remain under True Forward and retain Partial Commit on hardware support.
That makes early renewal a genuine lever, but it is only worth pulling if the pre-emptive term and pricing are otherwise competitive, since you are trading negotiating time for model preservation.
